Abstract: Background　Diagnosis and treatment delay is a kind of phenomenon of objective behavior experiencing a series of psychological activities.Understanding the psychology and causes of diagnosis and treatment delay of patients can effectively improve the effectiveness of their treatment，thereby promoting the continuation of patient health and the efficient allocation of medical and health resources.Objective　To compile the Diabetes Mellitus Diagnosis and Treatment Delayed Cognitive Behavioral Intention Scale （DMDBIS），and conduct reliability and validity tests.Methods　The pool of scale entries was formed through literature review combined with semi-structured interviews，based on the theory of planned behavior from January 2016 to February 2017.And then DMDBIS-Ⅱ was formed through expert assessment and pre-testing.280 diabetes patients with diagnosis and treatment delay were selected from the endocrine ward and outpatient clinics of three general hospitals including Daqing Oilfield General Hospital，the Fifth Hospital of Daqing City and the Fourth Hospital of Daqing as samples using the purposive sampling method，and DMDBIS was finally formed through large sample test.In addition，the SPSS 22.0 and Amos 21.0 statistical software were used for data analysis to test the reliability and validity of DMDBIS.Results　The DMDBIS includes 20 items in 5 dimensions，including namely symptom alertness （4 items），diabetes knowledge（6 items），habit reaction style（3 items），disease control（3 items） and factors of hindrance to medical treatment（4 items）.The results of DMDBIS exploratory factor analysis show that the KMO value was 0.810，and Bartlett spherical test χ2=3 305.296（df=190，P<0.001），which indicates that it is suitable for factor analysis.Principal component analysis shows that the cumulative variance contribution rate was 73.966%.Confirmatory factor analysis results show that the goodness-of-fit χ2=351.451，df=160，χ2/df=2.197，the approximation error root mean square （RMSEA）=0.069，the goodness of fit index（GFI）=0.882，comparison fitting index（CFI）=0.941，adjust the goodness of fit index （AGFI）=0.845，value-added adaptation index （IFI）=0.941，the gauge adapter index （TLI）=0.929，gauged adaptation index （NFI）=0.897.DMDBIS reliability tests showed that Cronbach's α coefficient of each factor ranged from 0.840 to 0.956，and Cronbach's α of the total table was 0.799.Conclusion　Collectively，the preliminary developed DMDBIS consists of 5 dimensions and 20 items with good reliability and validity.Therefore，it can provide a scientific measurement tool for diabetes prevention and treatment workers.

Key words: Diabetes mellitus, Diagnosis and treatment delay, Cognitive behavioral intention, Diabetes Mellitus Diagnosis and Treatment Delayed Cognitive Behavioral Intention Scale

摘要： 背景　诊治延误是一种经历一系列心理活动的客观行为现象。了解患者诊治延误的心理和原因可有效改善其治疗的效果，从而促进患者健康的延续和医疗卫生资源的高效配置。目的　编制糖尿病诊治延误认知行为意向量表（DMDBIS），并进行信效度检验。方法　2016年1月—2017年2月以计划行为理论为基础，结合文献回顾和通过半结构化访谈形成量表条目池，再通过专家评定、预测试形成DMDBIS暂定版量表（DMDBIS-Ⅱ）。采用目的抽样法选取大庆市油田总医院、大庆市第五医院及大庆市第四医院三所综合性医院内分泌病房及门诊已确诊为糖尿病且有诊治延误现象的280例患者为样本进行大样本测试，最终形成DMDBIS。采用SPSS 22.0及Amos 21.0统计学软件进行数据分析，检验DMDBIS的信效度。结果　最终形成的DMDBIS由5个维度20个条目组成，分别为症状警觉（4个条目）、糖尿病知识（6个条目）、习惯反应样式（3个条目）、疾病控制（3个条目）、阻碍就医因素（4个条目）。DMDBIS探索性因子分析结果显示：KMO值为0.810，Bartlett球形检验χ2=3 305.296（df=190，P<0.001），表明适合进行因子分析；主成分分析结果显示：累积方差贡献率为73.966%；验证性因子分析结果显示：拟合优度χ2=351.451、df=160、χ2/df=2.197，近似误差均方根（RMSEA）=0.069、拟合优度指数（GFI）=0.882、比较拟合指数（CFI）=0.941、调整拟合优度指数（AGFI）=0.845、增值适配指数（IFI）=0.941、非规准适配指数（TLI）=0.929、规准适配指数（NFI）=0.897。DMDBIS信度检验表明各因子的Cronbach's α系数为0.840~0.956，总量表的Cronbach's α系数为0.799。结论　总结而言，初步研制的DMDBIS共包括5个维度、20个条目，具有较好的信效度，可为糖尿病防治工作者提供科学的测量工具。

关键词: 糖尿病, 诊治延误, 认知行为意向, 糖尿病诊治延误认知行为意向量表